summaryrefslogtreecommitdiffstats
path: root/xlators
Commit message (Collapse)AuthorAgeFilesLines
* posix: Avoid one extra call of l(get|list)xattr call after use buffer in ↵Mohit Agrawal2017-07-032-177/+300
| | | | | | | | | | | | | | | | | | | | | posix_getxattr Problem: In posix xlator posix_(f)getxattr is calling system call(sys_lgetxattr) two times to fetch the xattr value. Solution: After use the extra buffer for first time calling we can avoid second attempt of system call(sys_lgetxattr) calling in posix_getxattr for most of the xattrs. BUG: 1460659 Change-Id: I0d8da776c5bc86653d874a4629a73bbf65c621b8 Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Reviewed-on: https://review.gluster.org/17520 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Kinglong Mee
* Link against missed libraries to resolve symbolsPrashanth Pai2017-07-033-4/+6
| | | | | | | | | | | | | | | | | | | | | | When external programs perform a dlopen("..so", RTLD_LAZY|RTLD_LOCAL) on some shared objects like xlators, it can fail with dlerror set to error string "undefined symbol <some-type>". This was observed for the following shared objects: fuse.so, quota.so, quotad.so, server.so, libgfrpc.so and socket.so P.S: This was found while running a go program which fetches the list of xlator options (volume_option_t) from xlator's shared object. BUG: 1193929 Change-Id: I7b958409cf11fb67c2be32a3f85a96fb1260236b Signed-off-by: Prashanth Pai <ppai@redhat.com> Reviewed-on: https://review.gluster.org/17659 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* multiple: fix struct/typedef inconsistenciesJeff Darcy2017-06-303-13/+13
| | | | | | | | | | | | | | | | | | | | | | | | The most common pattern, both in our code and elsewhere, is this: struct _xyz { ... }; typedef struct _xyz xyz_t; These exceptions - especially call_frame/call_stack - have been slowing down code navigation for years. By converging on a single pattern, navigating from xyz_t in code to the actual definition of struct _xyz (i.e. without having to visit the typedef first) might even be automatable. Change-Id: I0e5dd1f51f98e000173c62ef4ddc5b21d9ec44ed Signed-off-by: Jeff Darcy <jdarcy@fb.com> Reviewed-on: https://review.gluster.org/17650 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: Jeff Darcy <jeff@pl.atyp.us> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* features/shard: Remove ctx from LRU in shard_forgetPranith Kumar K2017-06-301-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: There is a race when the following two commands are executed on the mount in parallel from two different terminals on a sharded volume, which leads to use-after-free. Terminal-1: while true; do dd if=/dev/zero of=file1 bs=1M count=4; done Terminal-2: while true; do cat file1 > /dev/null; done In the normal case this is the life-cycle of a shard-inode 1) Shard is added to LRU when it is first looked-up 2) For every operation on the shard it is moved up in LRU 3) When "unlink of the shard"/"LRU limit is hit" happens it is removed from LRU But we are seeing a race where the inode stays in Shard LRU even after it is forgotten which leads to Use-after-free and then some memory-corruptions. These are the steps: 1) Shard is added to LRU when it is first looked-up 2) For every operation on the shard it is moved up in LRU Reader-handler Truncate-handler 1) Reader handler needs shard-x to be read. 1) Truncate has just deleted shard-x 2) In shard_common_resolve_shards(), it does inode_resolve() and that leads to a hit in LRU, so it is going to call __shard_update_shards_inode_list() to move the inode to top of LRU 2) shard-x gets unlinked from the itable and inode_forget(inode, 0) is called to make sure the inode can be purged upon last unref 3) when __shard_update_shards_inode_list() is called it finds that the inode is not in LRU so it adds it back to the LRU-list Both these operations complete and call inode_unref(shard-x) which leads to the inode getting freed and forgotten, even when it is in Shard LRU list. When more inodes are added to LRU, use-after-free will happen and it leads to undefined behaviors. Fix: I see that the inode can be removed from LRU even by the protocol layers like gfapi/gNFS when LRU limit is reached. So it is better to add a check in shard_forget() to remove itself from LRU list if it exists. BUG: 1466037 Change-Id: Ia79c0c5c9d5febc56c41ddb12b5daf03e5281638 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: https://review.gluster.org/17644 Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Krutika Dhananjay <kdhananj@redhat.com>
* cluster:dht Fix crash in dht_rename_lock_cbkN Balachandran2017-06-291-2/+4
| | | | | | | | | | | | | | | | | | Use a local variable to store the call count in the STACK_WIND for loop. Using frame->local is dangerous as it could be freed while the loop is still being processed Change-Id: Ie65cdcfb7868509b4a83bc2a5b5d6304eabfbc8e BUG: 1466110 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17645 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: Nigel Babu <nigelb@redhat.com> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: Check if fd is opened on dst subvolN Balachandran2017-06-286-30/+543
| | | | | | | | | | | | | | | | | | | | If an fd is opened on a file, the file is migrated and the cached subvol is updated in the inode_ctx before an fd based fop is sent, the fop is sent to the dst subvol on which the fd is not opened. This causes the FOP to fail with EBADF. Now, every fd based fop will check to see that the fd has been opened on the dst subvol before winding it down. Change-Id: Id92ef5eb7a5b5226688e2d2868b15e383f5f240e BUG: 1465075 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17630 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-by: Susant Palai <spalai@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* ec: Increase notification in all the casesAshish Pandey2017-06-281-31/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: "gluster v heal <volname> info" is taking long time to respond when a brick is down. RCA: Heal info command does virtual mount. EC wait for 10 seconds, before sending UP call to upper xlator, to get notification (DOWN or UP) from all the bricks. Currently, we are increasing ec->xl_notify_count based on the current status of the brick. So, if a DOWN event notification has come and brick is already down, we are not increasing ec->xl_notify_count in ec_handle_down. Solution: Handle DOWN even as notification irrespective of what is the current status of brick. Change-Id: I0acac0db7ec7622d4c0584692e88ad52f45a910f BUG: 1464091 Signed-off-by: Ashish Pandey <aspandey@redhat.com> Reviewed-on: https://review.gluster.org/17606 Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
* glusterd: mark brickinfo to started on successful attachAtin Mukherjee2017-06-281-5/+4
| | | | | | | | | | | | | | brickinfo's port & status should be filled up only when attach brick is successful. Change-Id: I68b181be37cb94d176f0f4692e8d9dac5493181c BUG: 1465559 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: https://review.gluster.org/17640 Reviewed-by: Jeff Darcy <jeff@pl.atyp.us> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Samikshan Bairagya <samikshan@gmail.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* glusterd: brick process fails to restart after gluster pod failureMohit Agrawal2017-06-271-10/+31
| | | | | | | | | | | | | | | | | | | | Problem: In container environment sometime after delete gluster pod and created new gluster pod brick process doesn't seem to come up. Solution: On the basis of logs it seems glusterd is trying to attach with non glusterfs process.Change the code of function glusterd_get_sock_from_brick_pid to fetch socketpath from argument of running brick process. BUG: 1464072 Change-Id: Ida6af00066341b683bbb4440d7a0d8042581656a Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Reviewed-on: https://review.gluster.org/17601 Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
* cluster/rebalance: Use GF_XATTR_LIST_NODE_UUIDS_KEY to figure out local subvols.Susant Palai2017-06-263-55/+54
| | | | | | | | | | | | | | | | | | | | | | Afr has introduced a new key GF_XATTR_LIST_NODE_UUIDS_KEY, through which rebalance will figure out its local subvolumes.(Reference bugid=1463250) key: GF_XATTR_NODE_UUID_KEY will continue to serve it's old purpose of returning the first afr chiild. test: prove tests/basic/distribute/rebal-all-nodes-migrate.t Change-Id: I4d602feda2a05b29d2210c712a07a4ac6b8bc112 BUG: 1463648 Signed-off-by: Susant Palai <spalai@redhat.com> Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17595 Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* cluster/dht: rebalance gets file count periodicallyN Balachandran2017-06-234-30/+97
| | | | | | | | | | | | | | | | The rebalance used to get the file count in the beginning and not update it. This caused estimates to fail if the number changed during the rebalance. The rebalance now updates the file count periodically. Change-Id: I1667ee69e8a1d7d6bc6bc2f060fad7f989d19ed4 BUG: 1464110 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17607 Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/ec: Node uuid xattr support update for ECSunil Kumar Acharya2017-06-232-6/+23
| | | | | | | | | | | | | | | | | | | | | | | | | Problem: The change in EC to return list of node uuids for GF_XATTR_NODE_UUID_KEY was causing problems with geo-rep. Fix: This patch will allow to get the single node uuid as it was doing before with the key "GF_XATTR_NODE_UUID_KEY", and will also allow to get the list of node uuids by using a new key "GF_XATTR_LIST_NODE_UUIDS_KEY". This will solve the problem with geo-rep and any other features which were depending on this. BUG: 1462790 Change-Id: I2d9214a9658d4a41a3d6de08600884d2bda5f3eb Signed-off-by: Sunil Kumar Acharya <sheggodu@redhat.com> Reviewed-on: https://review.gluster.org/17594 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* dht/hardlink : Remove stale linkto file incase of failureJiffin Tony Thottan2017-06-224-1/+84
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is a similar issue fixed for rename in https://review.gluster.org/#/c/16016/ For hardlinks, if cached and hashed subvolumes are different, then it will first create linkto file in hashed using root permission, but actually hardlink creation fails with EACESS and stale linkto file is never removed.All the followup hardlink calls with file name will result ESTALE because linktofile creation fails with EEXIST and follow up lookup on linkto file returns gfid-mismatching(old linkto file) and finally fails with ESTALE Steps to produce : (From link/00.t test from posix-testsuite) Steps executed in script * create a file "abc" using root * change the ownership of file to a non root user * create hardlink "link" for "abc" using a non root user, it fails with EACESS * delete "abc" * create directory "abc" using root * again try to create hadrlink "link" for "abc" using non root user, fails with ESTALE Also tried to fix other bugs in dht_linkfile_create_cbk() and posix_lookup. Thanks Susant for the help in debugging the issue and suggestion for this patch. Change-Id: I7a5a1899d3fd1fdb13578b37f9d52a084492e35d BUG: 1452084 Signed-off-by: Jiffin Tony Thottan <jthottan@redhat.com> Reviewed-on: https://review.gluster.org/17331 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: N Balachandran <nbalacha@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* feature/changelog: Fix buffer overflow crashKotresh HR2017-06-211-2/+4
| | | | | | | | | | | | | | | | The buffer used to hold the basename was hard coded to the size of NAME_MAX(255). It might lead to buffer overflow crashes when the basename which is sent is more than NAME_MAX length. Fixed the same. Change-Id: I6c1cad3ccaeb8c55549b1d3c5f96a198f65ba2b7 BUG: 1463178 Signed-off-by: Kotresh HR <khiremat@redhat.com> Reviewed-on: https://review.gluster.org/17579 CentOS-regression: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com>
* cluster/afr: Returning single and list of node uuids from AFRkarthik-us2017-06-201-9/+79
| | | | | | | | | | | | | | | | | | | | | | | | Problem: The change in afr to return list of node uuids was causing problems with geo-rep. Fix: This patch will allow to get the single node uuid as it was doing before with the key "GF_XATTR_NODE_UUID_KEY", and will also allow to get the list of node uuids by using a new key "GF_XATTR_LIST_NODE_UUIDS_KEY". This will solve the problem with geo-rep and any other feature which were depending on this. Change-Id: I09885dac6dfca127be94b708470c8c2941356f9a BUG: 1462790 Signed-off-by: karthik-us <ksubrahm@redhat.com> Reviewed-on: https://review.gluster.org/17576 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Ravishankar N <ravishankar@redhat.com> Reviewed-by: Kotresh HR <khiremat@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* dht: passing the errno as an argument to gf_msgAnkitRaj2017-06-201-110/+109
| | | | | | | | | | | | | | | | | | | There are many calls in gf_msg where errno is needed to pass as an argument instead of strerrno(error) Change-Id: I15048a5e0b41f9752a2023afe8470eca6f2cd383 Bug: 1454701 Signed-off-by: AnkitRaj <anraj@redhat.com> Reviewed-on: https://review.gluster.org/17464 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: MOHIT AGRAWAL <moagrawa@redhat.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> Tested-by: N Balachandran <nbalacha@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Karthik U S <ksubrahm@redhat.com> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* cluster/dht: Additional checks for rebalance estimatesN Balachandran2017-06-201-5/+15
| | | | | | | | | | | | | | | | | | | | | The rebalance estimates calculation was not handling calculations correctly when no files had been processed, i.e., when rate_lookedup was 0. Now, the estimated time is set to 0 in such scenarios as there is no way for rebalance to figure out how long the process will take to complete without knowing the rate at which the files are being processed. Change-Id: I7b6378e297e1ba139852bcb2239adf2477336b5b BUG: 1457985 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17564 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* posix: Revert modifying op_errno in __posix_fd_ctx_getRavishankar N2017-06-191-10/+6
| | | | | | | | | | | | | | | | | https://review.gluster.org/#/c/17414/ converted ENOENT to EBADFD because ENOENT is not a valid error for fd based operations, but this apparently breaks dht rebalance behaviour (see comments in the backport 17517. So reverting that part of the change. Change-Id: Idcf5c65a47b096a3766cf7f20ca938d988572052 BUG: 1456582 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: https://review.gluster.org/17565 Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* cluster/afr: Implement quorum for lk fopPranith Kumar K2017-06-192-23/+38
| | | | | | | | | | | | | | | | | | | Problem: At the moment when we have replica 3 or arbiter setup, even when lk succeeds on just one brick we give success to application which is wrong Fix: Consider quorum-number of successes as success when quorum is enabled. BUG: 1461792 Change-Id: I5789e6eb5defb68f8a0eb9cd594d316f5cdebaea Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: https://review.gluster.org/17524 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Ravishankar N <ravishankar@redhat.com>
* index: Do not proceed with init if brick is not mountedRavishankar N2017-06-194-4/+38
| | | | | | | | | | | | | | | | | | | | | ..or else when a volume start force is given, we end up creating /brick-path/.glusterfs/indices folder and various subdirs under it and eventually starting the brick process. As a part of this patch, glusterd_get_index_basepath() is added in glusterd, who will then use it to create the basepath during volume-create, add-brick, replace-brick and reset-brick. It also uses this function to set the 'index-base' xlator option for the index translator. Change-Id: Id018cf3cb6f1e2e35b5c4cf438d1e939025cb0fc BUG: 1457202 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: https://review.gluster.org/17426 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* cluster/ec: lk shouldn't be a transactionPranith Kumar K2017-06-161-19/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem: When application sends a blocking lock, the lk fop actually waits under inodelk. This can lead to a dead-lock. 1) Let's say app-1 takes exculsive-fcntl-lock on the file 2) app-2 attempts an exclusive-fcntl-lock on the file which goes to blocking stage note: app-2 is blocked inside transaction which holds an inode-lock 3) app-1 tries to perform write which needs inode-lock so it gets blocked on app-2 to unlock inodelk and app-2 is blocked on app-1 to unlock fcntl-lock Fix: Correct way to fix this issue and make fcntl locks perform well would be to introduce 2-phase locking for fcntl lock: 1) Implement a try-lock phase where locks xlator will not merge lk call with existing calls until a commit-lock phase. 2) If in try-lock phase we get quorum number of success without any EAGAIN error, then send a commit-lock which will merge locks. 3) In case there are any errors, unlock should just delete the lock-object which was tried earlier and shouldn't touch the committed locks. Unfortunately this is a sizeable feature and need to be thought through for any corner cases. Until then remove transaction from lk call. BUG: 1455049 Change-Id: I18a782903ba0eb43f1e6526fb0cf8c626c460159 Signed-off-by: Pranith Kumar K <pkarampu@redhat.com> Reviewed-on: https://review.gluster.org/17542 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Ashish Pandey <aspandey@redhat.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
* feature/bitrot: Fix ondemand scrubKotresh HR2017-06-162-6/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | The flag which keeps tracks of whether the scrub frequency is changed from previous value should not be considered for on-demand scrubbing. It should be considered only for 'scrub-frequency' where it should not be re-scheduled if it is set to same value again. But in case ondemand scrub, it should start the scrub immediately no matter what the scrub-frequency. Reproducer: 1. Enable bitrot 2. Set scrub-throttle 3. Set ondemand scrub Make sure glusterd is not restarted while doing below steps Change-Id: Ice5feaece7fff1579fb009d1a59d2b8292e23e0b BUG: 1461845 Signed-off-by: Kotresh HR <khiremat@redhat.com> Reviewed-on: https://review.gluster.org/17552 Smoke: Gluster Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra Bhat <raghavendra@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
* debug/io-stats: Provide option to select stats output formatKrutika Dhananjay2017-06-152-10/+45
| | | | | | | | | | | | | | ... as opposed to hardcoding it to "json" always. Change-Id: I5e79473a514373145ad764f24bb6219a6983a4c6 BUG: 1458197 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: https://review.gluster.org/17451 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Niels de Vos <ndevos@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com>
* glusterd: fix crash on statedump when no volumes are startedAtin Mukherjee2017-06-151-12/+17
| | | | | | | | | | | | | | | | | pmap object is created when glusterd allocates a port for the very first time, however before that if someone tries to take statedump glusterd will crash. Solution : Add a NULL check before accessing pmap reference. Change-Id: I206b02e07a4717e68af2c6bf05fac55119353de8 BUG: 1461655 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: https://review.gluster.org/17549 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Samikshan Bairagya <samikshan@gmail.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* glusterd: fix quorum calculation if percentage is not specified correctlyMichael Adam2017-06-141-1/+0
| | | | | | | | | | | | | | | | | | | | | | | There was an extra "ratio = _gf_true". - In case the ratio was specified correctly in the volfile, this is redundant. - In case the ratio was specified, but not parseable into a precentage, this is wrong and would lead to a quorum count of 0 instead of falling back to the default of 50% + 1. This patch removes the extra setting of "ratio". Change-Id: I2bd57ebf1b8989e905481a2b6285a1f422942f72 BUG: 1461129 Signed-off-by: Michael Adam <obnox@samba.org> Reviewed-on: https://review.gluster.org/17538 Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Tested-by: Atin Mukherjee <amukherj@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Smoke: Gluster Build System <jenkins@build.gluster.org>
* dht: reduce duplicate code in tier migrateAmar Tumballi2017-06-141-75/+10
| | | | | | | | | | | | | | | Updates #234 Change-Id: I016f6d4f1e5ad2ea56a611c1bffbd189f10650db Signed-off-by: Amar Tumballi <amarts@redhat.com> Reviewed-on: https://review.gluster.org/17525 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: hari gowtham <hari.gowtham005@gmail.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Susant Palai <spalai@redhat.com> Reviewed-by: N Balachandran <nbalacha@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* Revert "glusterd: disallow rebalance & remove-brick on a sharded volume"Krutika Dhananjay2017-06-132-19/+0
| | | | | | | | | | | | | | | | | This reverts commit 8375b3d70d5c6268c6770b42a18b2e1bc09e411e. Now that some of the users have confirmed rebalance works fine without causing corruption of VMs, time to revert the CLI restriction. Change-Id: I45493fcbb1f25fd0fff27b2b3526c42642ccb464 BUG: 1460585 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: https://review.gluster.org/17506 Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* nl-cache: Fix a possible crash and stale cachePoornima G2017-06-134-49/+141
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Issue1: Consider the followinf sequence of operations: ... nlc_ctx = nlc_ctx_get (inode i1) ....... -> nlc_clear_cache (i1) gets called as a part of nlc_invalidate or any other callers ... GF_FREE (ii nlc_ctx) LOCK (nlc_ctx->lock); -> This will result in crash as the ctx got freed in nlc_clear_cache. Issue2: lookup on dir1/file1 result in ENOENT add cache to dir1 at time T1 .... CHILD_DOWN at T2 lookup on dir1/file2 result in ENOENT add cache to dir1, but the cache time is still T1 lookup on dir1/file2 - should have been served from cache but the cache time is T1 < T2, hence cache is considered as invalid. So, after CHILD_DOWN the right thing would be to clear the cache and restart caching on that inode. Solution: Do not free nlc_ctx in nlc_clear_cache, but only in inode_forget() The fix for both issue1 and 2 is interleaved hence sending it as single patch. Change-Id: I83d8ed36c049a93567c6d7e63d045dc14ccbb397 BUG: 1458539 Signed-off-by: Poornima G <pgurusid@redhat.com> Reviewed-on: https://review.gluster.org/17453 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* readdir-ahead: Fix duplicate listing and cache size calculationPoornima G2017-06-121-19/+16
| | | | | | | | | | | | | | | | | | | | | | | Issue: If a opendir is followed by a closedir without readdir, though the prefetched entries were freed, the freed size was not accounted in priv->rda_cache_size. Thus the cache limit will exceed if there are multiple opendir followed by closedir. Fix: Fix the pric->rda_cache_size calculation. Also have removed the inode_ctx_size. Each perf xlator has its own cache limit that it works with. Also the inode_ctx size can change, if a forget/ invalidate or any other factor triggers the inode_ctx size. Change-Id: I9707ec558076ce046e58a55989ec9513c70ea029 BUG: 1431908 Signed-off-by: Poornima G <pgurusid@redhat.com> Reviewed-on: https://review.gluster.org/17504 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* md-cache: Fix the dump of stat inode in .meta and statedumpPoornima G2017-06-121-8/+8
| | | | | | | | | | Change-Id: If61ed5e4462e98d18a1370734a0bcee1ed94d82d Signed-off-by: Poornima G <pgurusid@redhat.com> Reviewed-on: https://review.gluster.org/17491 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* glusterd: log stale rpc disconnects occasionallyAtin Mukherjee2017-06-091-3/+6
| | | | | | | | | | | | | | | | There might be situations where if a brick process is killed through SIGKILL (not SIGTERM) when brick mux is enabled glusterd will continue to receive disconnect events from the stale rpc which might flood the glusterd log. Fix is to use GF_LOG_OCCASIONALLY. Change-Id: I95a10c8be2346614e0a3458f98d9f99aab34800a BUG: 1460225 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: https://review.gluster.org/17499 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* glusterfsd: Deletion of brick dir throw emerg msgs after stop volumeMohit Agrawal2017-06-093-5/+6
| | | | | | | | | | | | | | | | | | | | Problem: Deletion of brick directories throw emerg messages after stop volume while brick mux is enabled. Solution: Modify the posix health check monitor thread code to handled correctly. BUG: 1459781 Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Change-Id: I2d22a84f9a98b0da261e5fb7850ba1368f3601d7 Reviewed-on: https://review.gluster.org/17492 Tested-by: MOHIT AGRAWAL <moagrawa@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* upcall: Update the access time in missing casesPoornima G2017-06-093-22/+53
| | | | | | | | | | | | | | | | | | Issue: In fops like rename, link, unlink etc, the parent dirrs' client access time was not being updated. And in fops like create, link, symlink etc. the new file/dirs' client access time was not updated. Solution: Update the client access time for both parent and new entry. Change-Id: Id9f63583216ae857f6251dca15797ac66fa85430 BUG: 1458127 Signed-off-by: Poornima G <pgurusid@redhat.com> Reviewed-on: https://review.gluster.org/17450 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* posix-acl: Whitelist virtual ACL xattrsSoumya Koduri2017-06-091-0/+4
| | | | | | | | | | | | | | | | Similar to system.posix_acl_* xattrs, all users should get permission to be able to read glusterfs.posix.acl* xattrs too. Change-Id: I1fc2b67c8a12113910e4ec57cd114e4baefe0d38 BUG: 1459971 Signed-off-by: Soumya Koduri <skoduri@redhat.com> Reviewed-on: https://review.gluster.org/17493 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: jiffin tony Thottan <jthottan@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com>
* debug/trace: Implement seek fopKrutika Dhananjay2017-06-091-0/+46
| | | | | | | | | | | Change-Id: Idd1b92201348a76a6145b38799a3a46cab730540 BUG: 1458193 Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com> Reviewed-on: https://review.gluster.org/17437 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Niels de Vos <ndevos@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* protocol/server: make listen backlog value as configurableMohammed Rafi KC2017-06-085-7/+72
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | problem: When we call listen from protocol/server, we are giving a hard coded valie of 10 if it is not manually given. With multiplexing, especially when glusterd restarts all clients may try to connect to the server at a time. Which will result in overflowing the queue, and kernel will complain about the errors. Solution: This patch will introduce a volume set command to make backlog value as a configurable. This patch also changes the default values for backlog from 10 to 128. This changes is only applicable for sockets listening from protocol. Example: gluster volume set <volname> transport.listen-backlog 1024 Note: 1 Brick has to be restarted to get this value in effect 2 This changes won't be reflected in glusterd, or other xlators which calls listen. If you need, you have to add this option to the volfile. Change-Id: I0c5a2bbf28b5db612f9979e7560e05dd82b41477 BUG: 1456405 Signed-off-by: Mohammed Rafi KC <rkavunga@redhat.com> Reviewed-on: https://review.gluster.org/17411 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-by: Niels de Vos <ndevos@redhat.com> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* glusterfsd: Sometime brick process is crashed after enable brick muxMohit Agrawal2017-06-072-6/+19
| | | | | | | | | | | | | | | | | Problem: glusterfsd is getting Segfault while running bug-1432542-mpx-restart-crash.t in a loop in while brick mux is enabled. Solution: Change the index_worker code as well as notify code in index.c to cleanup index_worker thread appropriately. BUG: 1459402 Change-Id: Id036664dc34309dd3f6e54746fc2724182cb074f Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Reviewed-on: https://review.gluster.org/17471 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* glusterd: fix glusterd crash from glusterd_op_ac_rcvd_brick_op_accAtin Mukherjee2017-06-071-1/+1
| | | | | | | | | | | | | | | In out label, before checking ev_ctx->rsp_dict we should first check if ev_ctx is not NULL Change-Id: I28f4f1ee9070617a0e6a23a43af8c5756f96a47e BUG: 1452956 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: https://review.gluster.org/17478 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Gaurav Yadav <gyadav@redhat.com> Reviewed-by: Samikshan Bairagya <samikshan@gmail.com>
* cluster/dht: Include dirs in rebalance estimatesN Balachandran2017-06-073-31/+83
| | | | | | | | | | | | | | | | | Empty directories were not being considered while calculating rebalance estimates leading to negative time-left values being displayed as part of the rebalance status. Change-Id: I48d41d702e72db30af10e6b87b628baa605afa98 BUG: 1457985 Signed-off-by: N Balachandran <nbalacha@redhat.com> Reviewed-on: https://review.gluster.org/17448 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* cluster/ec: Update xattr and heal size properlyAshish Pandey2017-06-062-9/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Problem-1 : Recursive healing of same file is happening when IO is going on even after data heal completes. Solution: RCA: At the end of the write, when ec_update_size_version gets called, we send it only on good bricks and not on healing brick. Due to this, xattr on healing brick will always remain out of sync and when the background heal check source and sink, it finds this brick to be healed and start healing from scratch. That involve ftruncate and writing all of the data again. To solve this, send xattrop on all the good bricks as well as healing bricks. Problem-2: The above fix exposes the data corruption during heal. If the write on a file is going on and heal finishes, we find that the file gets corrupted. RCA: The real problem happens in ec_rebuild_data(). Here we receive the 'size' argument which contains the real file size at the time of starting self-heal and it's assigned to heal->total_size. After that, a sequence of calls to ec_sync_heal_block() are done. Each call ends up calling ec_manager_heal_block(), which does the actual work of healing a block. First a lock on the inode is taken in state EC_STATE_INIT using ec_heal_inodelk(). When the lock is acquired, ec_heal_lock_cbk() is called. This function calls ec_set_inode_size() to store the real size of the inode (it uses heal->total_size). The next step is to read the block to be healed. This is done using a regular ec_readv(). One of the things this call does is to trim the returned size if the file is smaller than the requested size. In our case, when we read the last block of a file whose size was = 512 mod 1024 at the time of starting self-heal, ec_readv() will return only the first 512 bytes, not the whole 1024 bytes. This isn't a problem since the following ec_writev() sent from the heal code only attempts to write the amount of data read, so it shouldn't modify the remaining 512 bytes. However ec_writev() also checks the file size. If we are writing the last block of the file (determined by the size stored on the inode that we have set to heal->total_size), any data beyond the (imposed) end of file will be cleared with 0's. This causes the 512 bytes after the heal->total_size to be cleared. Since the file was written after heal started, the these bytes contained data, so the block written to the damaged brick will be incorrect. Solution: Align heal->total_size to a multiple of the stripe size. Thanks "Xavier Hernandez" <xhernandez@datalab.es> to find out the root cause and to fix the issue. Change-Id: I6c9f37b3ff9dd7f5dc1858ad6f9845c05b4e204e BUG: 1428673 Signed-off-by: Ashish Pandey <aspandey@redhat.com> Reviewed-on: https://review.gluster.org/16985 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Reviewed-by: Xavier Hernandez <xhernandez@datalab.es>
* glusterd: fix brick start raceAtin Mukherjee2017-06-065-21/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit tries to handle a race where we might end up trying to spawn the brick process twice with two different set of ports resulting into glusterd portmapper having the same brick entry in two different ports which will result into clients to fail connect to bricks because of incorrect ports been communicated back by glusterd. In glusterd_brick_start () checking brickinfo->status flag to identify whether a brick has been started by glusterd or not is not sufficient as there might be cases where while glusterd restarts glusterd_restart_bricks () will be called through glusterd_spawn_daemons () in synctask and immediately glusterd_do_volume_quorum_action () with server-side-quorum set to on will again try to start the brick and in case if the RPC_CLNT_CONNECT event for the same brick hasn't been processed by glusterd by that time, brickinfo->status will still be marked as GF_BRICK_STOPPED resulting into a reattempt to start the brick with a different port and that would result portmap go for a toss and resulting clients to fetch incorrect port. Fix would be to introduce another enum value called GF_BRICK_STARTING in brickinfo->status which will be set when a brick start is attempted by glusterd and will be set to started through RPC_CLNT_CONNECT event. For brick multiplexing, on attach brick request given the brickinfo->status flag is marked to started directly this value will not have any effect. Also this patch removes started_here flag as it looks to be redundant as brickinfo->status. Change-Id: I9dda1a9a531b67734a6e8c7619677867b520dcb2 BUG: 1457981 Signed-off-by: Atin Mukherjee <amukherj@redhat.com> Reviewed-on: https://review.gluster.org/17447 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* dht: for many operation directory/file path is (null) in brick logAnkitRaj2017-06-051-4/+8
| | | | | | | | | | | | | | | | There are few operations on mount point which results to (null) error for path on brick log. This should be corrected with correct path. Change-Id: I5acf2ead5196594c76a2e526aafbfc128ece433e BUG: 1456361 Signed-off-by: ankitraj <anraj@redhat.com> Reviewed-on: https://review.gluster.org/17407 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: ankitraj NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra G <rgowdapp@redhat.com>
* afr: update errno check in afr_inode_refresh_doRavishankar N2017-06-051-1/+1
| | | | | | | | | | | | | Addresses review comment in https://review.gluster.org/#/c/17413 Change-Id: Ic247729e5e92a5bb0148543764e0b30790444004 BUG: 1456582 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: https://review.gluster.org/17436 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
* core: fix spelling errorsKaleb S. KEITHLEY2017-06-026-9/+8
| | | | | | | | | | | | | | fixes for various minor spelling errors and typos Reported-by: Patrick Matthäi <pmatthaei@debian.org> Change-Id: Ic1be36f82e3d822bbdc9559878bd79520fc0fcd5 BUG: 1457808 Signed-off-by: Kaleb S. KEITHLEY <kkeithle@redhat.com> Reviewed-on: https://review.gluster.org/17442 NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Niels de Vos <ndevos@redhat.com> Smoke: Gluster Build System <jenkins@build.gluster.org>
* glusterd: Fix regression wrt add-brick on replica count changeSamikshan Bairagya2017-06-012-3/+11
| | | | | | | | | | | | | | | | | | | | | | | | | tests/bugs/glusterd/bug-1406411-fail-add-brick-on-replica-count-change.t was failing on centos machines with brick multiplexing enabled. This is because detaching individual bricks manually from the backend like it is done in the regression test framework by 'kill_brick', fails to send a RPC_CLNT_DISCONNECT to glusterd when multiplexing is enabled. This causes the add-brick command to not fail when one of the bricks are killed using kill_brick in the regression test framework. To fix this, set the brick status to GF_BRICK_STOPPED on the glusterd end during portmap signout. This commit also sets the brick status in glusterd_brick_stop() function so that the brick status is correctly set to 'stopped' even when the function is called independently for individual bricks. Change-Id: I4d6f7b579069d0cfa53cb2b0cff78876e1f31594 BUG: 1456898 Signed-off-by: Samikshan Bairagya <samikshan@gmail.com> Reviewed-on: https://review.gluster.org/17422 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us> Reviewed-by: Atin Mukherjee <amukherj@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* glusterfs: Not able to mount running volume after enable brick mux and ↵Mohit Agrawal2017-05-3116-41/+261
| | | | | | | | | | | | | | | | | | | | | | | | | | | | stopped any volume Problem: After enabled brick mux if any volume has down and then try ot run mount with running volume , mount command is hung. Solution: After enable brick mux server has shared one data structure server_conf for all associated subvolumes.After down any subvolume in some ungraceful manner (remove brick directory) posix xlator sends GF_EVENT_CHILD_DOWN event to parent xlatros and server notify updates the child_up to false in server_conf.When client is trying to communicate with server through mount it checks conf->child_up and it is FALSE so it throws message "translator are not yet ready". From this patch updated structure server_conf to save child_up status for xlator wise. Another improtant correction from this patch is cleanup threads from server side xlators after stop the volume. BUG: 1453977 Change-Id: Ic54da3f01881b7c9429ce92cc569236eb1d43e0d Signed-off-by: Mohit Agrawal <moagrawa@redhat.com> Reviewed-on: https://review.gluster.org/17356 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Raghavendra Talur <rtalur@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Jeff Darcy <jeff@pl.atyp.us>
* posix: use the correct op_errnoRavishankar N2017-05-312-8/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | Problem: If readdir/fstat was performed on a directory that was removed, posix_fd_ctx_get() fails with ENOENT but we incorrectly use the ret value (-1 in this case) as op_errno, logging "Operation not permitted" messages in the brick logs. Also in case of fstat, the -1 op_errno was also propagated to the client via stack unwind, causing the message to appear in protocol/client logs as well. Fix: Use the right op_errno in readdir, fstat and writev. Also, if posix_fd_ctx_get() failed with ENOENT, convert it into EBADF because ENOENT is not a valid error for an fd operation. Change-Id: Ie43c0789d5040ec73b7cf885d015a183b8c64d70 BUG: 1456582 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: https://review.gluster.org/17414 Smoke: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com> Tested-by: Pranith Kumar Karampuri <pkarampu@redhat.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
* cluster/dht: Make optimal usage of buffer provided with readdir(p)Sakshi2017-05-314-54/+62
| | | | | | | | | | | | | | | | | | | dht_readdirp must unwind with list of entries only after the entire buffer requested by kernel is filled to avoid extra syscalls occuring when returning partially filled buffer. Also wind readdir call to next subvol on reaching EOD for directory on that subvol to avoid extra network call. Change-Id: If2e1a2722f813d95457c7542bff25fef56c7a041 BUG: 1356453 Signed-off-by: Sakshi <sabansal@redhat.com> Signed-off-by: Raghavendra G <rgowdapp@redhat.com> Reviewed-on: https://review.gluster.org/12271 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Amar Tumballi <amarts@redhat.com> Reviewed-by: Susant Palai <spalai@redhat.com>
* tier: port value missing on cli parsinghari gowtham2017-05-311-0/+10
| | | | | | | | | | | | | | | | | | | | | problem: as tier didn't have a port all the values regarding the port were removed. but the cli needs a port value to parse and print the status. fix: fake the port value with a zero. Change-Id: I6491f6c441f7cfddbdaa724fcbe7c30e348aa765 BUG: 1452006 Signed-off-by: hari gowtham <hgowtham@redhat.com> Reviewed-on: https://review.gluster.org/17419 Smoke: Gluster Build System <jenkins@build.gluster.org> Tested-by: hari gowtham <hari.gowtham005@gmail.com> Reviewed-by: Shyamsundar Ranganathan <srangana@redhat.com> Reviewed-by: Prashanth Pai <ppai@redhat.com> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Samikshan Bairagya <samikshan@gmail.com> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> Reviewed-by: Atin Mukherjee <amukherj@redhat.com>
* afr: add errno to afr_inode_refresh_done()Ravishankar N2017-05-311-7/+16
| | | | | | | | | | | | | | | | | | | | | | | | | Problem: When parellel `rm -rf`s were being done from cifs clients, opendir might fail on some replicas with ENOENT. DHT ignores partial opendir failures in dht_fd_cbk() and winds readdirs on those replicas. Afr inode refresh (as a part of readdirp read_txn) sees in its fd context that the state of the fds is *not* AFR_FD_OPENED and bails out to afr_inode_refresh_done() without doing a refresh. When this happens, the errno is set as EIO due to lack of readable subvols, logging split-brain messages in the logs. Fix: Introduce an errno argument to afr_inode_refresh_do() to bail out with the right error value when inode refresh is not performed. Change-Id: I075707fbb73fd93a923b77b923a96aac79e847f9 BUG: 1456582 Signed-off-by: Ravishankar N <ravishankar@redhat.com> Reviewed-on: https://review.gluster.org/17413 Smoke: Gluster Build System <jenkins@build.gluster.org> NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org> CentOS-regression: Gluster Build System <jenkins@build.gluster.org> Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
generated by cgit (git 2.34.1) at 2024-05-19 11:24:08 +0000